The following work was accomplished in 2011: Expression, Purification and Complex Formation of Bacterial and Human Proteins We solved the structure of full-length, apo human transferrin in 2007 and published this structure (Wally et al., J. Biol. Chem. 2007). Thispast year we used the information to phase a complex structure for TbpA-hTf described below, and at the same time we solved a much higher resolution structure of apo hTf C-lobe to 1.7 resolution. We developed an expression system for TbpA that is capable of producing 0.2-0.4 mg/L of purified protein. A large scale preparation (27L) of E. coli transformed with a plasmid containing the gene for TbpA as well as an N-terminal His-Tag is grown and induced with low level IPTG overnight and then harvested. We recently improved expression yields through targeted mutagenesis of the C-terminal beta strand, allowing the newly synthesized protein to be more efficiently processed by the E. coli membrane insertion machinery. We developed expression system for TbpB that is capable of producing 7-8 mg/L of purified protein. A medium scale preparation (4L) of E. coli transformed with a plasmid containing the gene for TbpB (a soluble construct lacking the N-terminal membrane anchor) as well as an N-terminal His-Tag and a cleavage site are grown and induced with high level IPTG for a few hours and then harvested. The cells are then lysed and membranes are separated via ultracentrifugation, and the soluble fraction is run over a Nickel-NTA column. The nickel eluate is then further purified and can be bound to purified commercially available iron-bound human serum transferrin. We solved the structure of TbpB to 2.65 resolution this year. Using our purified TbpB, which only binds holo hTf, we made a TbpB-holo hTf complex and characterized it using small angle X-ray scattering. Since we have structures of the two protein components, we were able to structurally characterize the complex, even though we have not been able to crystallize it. We recently obtained crystals of TbpA+hTf (180 kDa total) that diffract X-rays to 2.65 resolution. The structure was solved in August 2010 by molecular replacement using our hTf coordinates (Wally et al., JBC 2007) and various TonB-dependent transporter models. This structure, the first of an outer membrane protein in complex with its full-length host target (exemplifying host pathogen interactions), tells us about the specificity for human transferrin, the residues involved in this interaction, where variable and conserved epitopes are located on the extracellular surface (useful for vaccine design), how iron is extracted from hTf, and how it is transported across the outer membrane. When TbpA binds hTf, it sequesters 2500 2 of buried surface through an extensive interface consisting of 81 TbpA residues and 67 hTf residues. This structure is a landmark achievement, taking 12 years to complete. Recently, we made the first ever in vitro complex of TbpA-TbpB bound to holo human transferrin (hTf) and this triple complex was characterized by negative stain single particle electron microscopy. From the TbpA-hTf crystal structure and the TbpB-hTf SAXS structure, we made a model for the triple complex which fits beautifully into the EM 3D reconstruction. From all of these data, we now understand the specificity of recognition and have already designed antibodies that can block the TbpA-transferrin interaction. Since TbpA and TbpB are present in all clinical isolates, they represent important therapeutic targets. This is one of the largest outer membrane protein complexes characterized at high resolution and it has direct translational applications. A manuscript describing the crystallography, SAXS, EM, and functional analysis was recently submitted to Nature.